Physiological responses to violent game events: Does it matter whose character you kill?

1
Physiological responses to violent game events
Does it matter whose character you kill?
THE CENTER FOR KNOWLEDGE AND INNOVATION
RESEARCH H E L S I N K I S C H O O L O F
E C O N O M I C S

• J Matias Kivikangas, M.A. Niklas Ravaja, Ph.D.
• CKIR, Helsinki School of Economics, Finland
• email ravaja_at_hse.fi

2
Background

• Previous studies have shown that different
  violent and nonviolent game events elicit
  differential phasic emotion-related
  psychophysiological responses
• Ravaja, N., Saari, T., Salminen, M., Laarni, J.,
  Kallinen, K. (2006). Phasic emotional reactions
  to video game events A psychophysiological
  investigation. Media Psychology, 8, 343-367.
• Ravaja, N., Turpeinen, M., Saari, T., Puttonen,
  S., Keltikangas-Järvinen, L. (in revision). The
  Psychophysiology of James Bond Phasic Emotional
  Responses to Violent Video Game Events. Emotion.
• Previous studies have also shown that the type of
  opponent (computer, stranger, friend) influences
  emotional responses as indexed by tonic
  psychophysiological measures (mean values during
  the game session)
• Ravaja, N., Saari, T., Turpeinen, M., Laarni, J.,
  Salminen, M., Kivikangas, M. (2006). Spatial
  presence and emotions during video game playing
  Does it matter with whom you play? Presence
  Teleoperators and Virtual Environments, 15,
  381-392.
• A question arises
• Does it matter whose character you kill?
• Do phasic emotion-related physiological responses
  to violent events differ as a function of
  opponent type (computer, stranger, friend)?

3
Methods

• Participants
• Participants were 99 (51 male and 48 female)
  Finnish undergraduates, who ranged from 19 to 34
  years of age (mean 23.8 years)
• Participants participated in the experiment in
  groups of three same-sex persons. In each of the
  33 groups, two of the participants were friends
  who knew each other before and one was a person
  unknown to the others (i.e., a stranger).
• In the present study, we used only the
  self-report and physiological data collected
• from the 33 so-called main participants.
• Game
• Super Monkey Ball Jr. (Sega Corporation, Tokyo,
  Japan) and Duke Nukem Advance (Take 2
  Interactive, Berkshire, UK). The games were
  played with the Nintendo Game Boy Advance console
  (Nintendo Co., Ltd., Kyoto, Japan).
• Design
• A 2 (Game Super Monkey Ball Jr., Duke Nukem
  Advance) ? 3 (Opponent computer, stranger,
  friend) within-subjects design was employed

4
Methods

• Physiological measures
• Facial electromyography (EMG)
• Provides a direct measure of the electrical
  activity associated with facial muscle
  contractions (an important form of emotional
  expression)
• Zygomaticus major (cheek) muscle area
• An index of Positive Affect (PA)
• Orbicularis oculi (periocular) muscle area
• An index of PA
• Corrugator supercilii (brow) muscle area
• An index of Negative Affect (NA)
• Electrodermal activity (EDA)
• The primary psychophysiological index of arousal
• As people experience arousal their sympathetic
  nervous system is activated, resulting in
  increased sweat gland activity and skin
  conductance

5
Dimensional Model of Emotion
6
Methods

• Game Events
• The exact onset times of predefined game events
  were determined by examining the played games,
  frame by frame, using V-ToolsPro 2.20 software.
• We scored four different game events
• The players character (own PC) wounded an
  opponent character (NPC/PC Opponent Wounded)
• The players character (own PC) killed an
  opponent character (NPC/PC Opponent Killed)
• The players character (own PC) was wounded by an
  opponent character (NPC/PC Own PC Wounded)
• The players character (own PC) was killed by an
  opponent character (NPC/PC Own PC Killed)

7
Methods

• Data Analysis
• Mean values for the psychophysiological measures
  were derived for one 1-s epoch before each event
  (Second 1) and for six 1-s epochs after event
  onset (Seconds 2 to 7). The data were analyzed by
  the Linear Mixed Models procedure in SPSS with
  restricted maximum likelihood estimation and a
  first-order autoregressive covariance structure
  for the residuals.
• Event-related changes in physiological activity
  were tested using the following orthogonal
  contrasts
• Linear trend across seconds 1 to 7 (Contrast 1a)
• Quadratic trend across seconds 1 to 7 (Contrast
  1b)
• Friend vs. Stranger Linear Trend across Seconds
  1 to 7 (Contrast 2a)
• Friend vs. Stranger Quadratic Trend across
  Seconds 1 to 7 (Contrast 2b)
• Computer vs. Friend and Stranger Linear Trend
  across Seconds 1 to 7 (Contrast 3a)
• Computer vs. Friend and Stranger Quadratic
  Trend across Seconds 1 to 7 (Contrast 3b)

8
Hypotheses and Results I

• H1 The wounding and death of the players own
  character will elicit PA as indexed by increased
  zygomatic and orbicularis oculi EMG activity.

Event Own PC killed
Event Own PC wounded
Contrast 1a, p lt .001 Contrast 3a, p lt .001
Contrast 1b, p lt .001
9
Hypotheses and Results II

• H2 Killing or wounding the PC of another human
  (friend or stranger) will elicit greater PA as
  indexed by zygomatic and orbicularis oculi EMG
  activity compared to killing or wounding a NPC.

Event Opponent Killed
Event Opponent Wounded
Contrast 3b, p lt .001
Contrast 3a, p lt .001
10
Hypotheses and Results III

• H3 NA as indexed by corrugator EMG activity will
  increase in the order of killing or wounding a
  NPC lt killing or wounding the PC of a stranger lt
  killing or wounding the PC of a friend.

Event Opponent Killed
Event Opponent Wounded
Contrast 2b, p lt .001 Contrast 3a, p lt .001
Contrast 3b, p lt .001
Contrast 2b, p lt .05 Contrast 3a, p lt .001
11
Hypotheses and Results IV

• H4 Physiological arousal as indexed by EDA will
  increase in the order of killing or wounding a
  NPC lt killing or wounding the PC of a stranger lt
  killing or wounding the PC of a friend.
• H5 Physiological arousal as indexed by EDA will
  increase in the order of the players character
  is wounded or killed by a NPC lt the players
  character is wounded or killed by the PC of a
  stranger lt the players character is wounded or
  killed by the PC of a friend.

Event Own PC Wounded
Event Opponent Wounded
Contrast 3a, p lt .001
Contrast 2a, p lt .001 Contrast 3a, p lt .001
12
Conclusions

• Although counterintuitive, the wounding and death
  of the players own character elicits some aspect
  of PA
• Killing and wounding the PC of another human
  elicits both greater PA and NA compared to
  killing and wounding a NPC
• Arousal associated with all types of violent game
  events increases in the order of playing against
  a computer lt playing against a stranger lt playing
  against a friend